Analysis Airbnb

import libraries

import plotly.offline as pyo
# Set notebook mode to work in offline
pyo.init_notebook_mode()


import pandas as pd
import matplotlib.pyplot as plt

import seaborn as sns
import plotly.express as px
import geopandas as gpd
import pprint

import numpy as np
#plt.rcParams['figure.figsize'] = [20, 15]

Load listing data.

df_listings_detailed = pd.read_csv("./data_airbnb_syd/20211106/listings_detailed.csv", 
                                       header='infer', 
                                       parse_dates=['last_scraped', 'host_since', 'calendar_last_scraped'])
df_listings_detailed.head(2)

	id	listing_url	scrape_id	last_scraped	name	description	neighborhood_overview	picture_url	host_id	host_url	...	review_scores_communication	review_scores_location	review_scores_value	license	instant_bookable	calculated_host_listings_count	calculated_host_listings_count_entire_homes	calculated_host_listings_count_private_rooms	calculated_host_listings_count_shared_rooms	reviews_per_month
0	11156	https://www.airbnb.com/rooms/11156	20211106153124	2021-11-07	An Oasis in the City	Very central to the city which can be reached ...	It is very close to everything and everywhere,...	https://a0.muscache.com/pictures/2797669/17895...	40855	https://www.airbnb.com/users/show/40855	...	4.77	4.82	4.71	NaN	f	1	0	1	0	2.02
1	14250	https://www.airbnb.com/rooms/14250	20211106153124	2021-11-07	Manly Harbour House	Beautifully renovated, spacious and quiet, our...	Balgowlah Heights is one of the most prestigio...	https://a0.muscache.com/pictures/56935671/fdb8...	55948	https://www.airbnb.com/users/show/55948	...	4.33	4.67	4.33	Exempt	f	2	2	0	0	0.04

2 rows × 74 columns

Check the dataset's size and columns.

pprint.pprint(df_listings_detailed.shape)
pprint.pprint(df_listings_detailed.info())

(20513, 74)
<class 'pandas.core.frame.DataFrame'>
RangeIndex: 20513 entries, 0 to 20512
Data columns (total 74 columns):
 #   Column                                        Non-Null Count  Dtype         
---  ------                                        --------------  -----         
 0   id                                            20513 non-null  int64         
 1   listing_url                                   20513 non-null  object        
 2   scrape_id                                     20513 non-null  int64         
 3   last_scraped                                  20513 non-null  datetime64[ns]
 4   name                                          20506 non-null  object        
 5   description                                   19741 non-null  object        
 6   neighborhood_overview                         12365 non-null  object        
 7   picture_url                                   20513 non-null  object        
 8   host_id                                       20513 non-null  int64         
 9   host_url                                      20513 non-null  object        
 10  host_name                                     20508 non-null  object        
 11  host_since                                    20508 non-null  datetime64[ns]
 12  host_location                                 20493 non-null  object        
 13  host_about                                    10978 non-null  object        
 14  host_response_time                            7217 non-null   object        
 15  host_response_rate                            7217 non-null   object        
 16  host_acceptance_rate                          8565 non-null   object        
 17  host_is_superhost                             20508 non-null  object        
 18  host_thumbnail_url                            20508 non-null  object        
 19  host_picture_url                              20508 non-null  object        
 20  host_neighbourhood                            11738 non-null  object        
 21  host_listings_count                           20508 non-null  float64       
 22  host_total_listings_count                     20508 non-null  float64       
 23  host_verifications                            20513 non-null  object        
 24  host_has_profile_pic                          20508 non-null  object        
 25  host_identity_verified                        20508 non-null  object        
 26  neighbourhood                                 12367 non-null  object        
 27  neighbourhood_cleansed                        20513 non-null  object        
 28  neighbourhood_group_cleansed                  0 non-null      float64       
 29  latitude                                      20513 non-null  float64       
 30  longitude                                     20513 non-null  float64       
 31  property_type                                 20513 non-null  object        
 32  room_type                                     20513 non-null  object        
 33  accommodates                                  20513 non-null  int64         
 34  bathrooms                                     0 non-null      float64       
 35  bathrooms_text                                20489 non-null  object        
 36  bedrooms                                      19078 non-null  float64       
 37  beds                                          19646 non-null  float64       
 38  amenities                                     20513 non-null  object        
 39  price                                         20513 non-null  object        
 40  minimum_nights                                20513 non-null  int64         
 41  maximum_nights                                20513 non-null  int64         
 42  minimum_minimum_nights                        20513 non-null  int64         
 43  maximum_minimum_nights                        20513 non-null  int64         
 44  minimum_maximum_nights                        20513 non-null  int64         
 45  maximum_maximum_nights                        20513 non-null  int64         
 46  minimum_nights_avg_ntm                        20513 non-null  float64       
 47  maximum_nights_avg_ntm                        20513 non-null  float64       
 48  calendar_updated                              0 non-null      float64       
 49  has_availability                              20513 non-null  object        
 50  availability_30                               20513 non-null  int64         
 51  availability_60                               20513 non-null  int64         
 52  availability_90                               20513 non-null  int64         
 53  availability_365                              20513 non-null  int64         
 54  calendar_last_scraped                         20513 non-null  datetime64[ns]
 55  number_of_reviews                             20513 non-null  int64         
 56  number_of_reviews_ltm                         20513 non-null  int64         
 57  number_of_reviews_l30d                        20513 non-null  int64         
 58  first_review                                  14878 non-null  object        
 59  last_review                                   14878 non-null  object        
 60  review_scores_rating                          14878 non-null  float64       
 61  review_scores_accuracy                        14260 non-null  float64       
 62  review_scores_cleanliness                     14270 non-null  float64       
 63  review_scores_checkin                         14253 non-null  float64       
 64  review_scores_communication                   14270 non-null  float64       
 65  review_scores_location                        14254 non-null  float64       
 66  review_scores_value                           14249 non-null  float64       
 67  license                                       6761 non-null   object        
 68  instant_bookable                              20513 non-null  object        
 69  calculated_host_listings_count                20513 non-null  int64         
 70  calculated_host_listings_count_entire_homes   20513 non-null  int64         
 71  calculated_host_listings_count_private_rooms  20513 non-null  int64         
 72  calculated_host_listings_count_shared_rooms   20513 non-null  int64         
 73  reviews_per_month                             14878 non-null  float64       
dtypes: datetime64[ns](3), float64(19), int64(21), object(31)
memory usage: 11.6+ MB
None

This dataset has 78 columns. Have a look at one row to check what infomation is present.

i = 0
for col, val in df_listings_detailed.iloc[0].iteritems():
    print(i, '\t', col, '\t', val)
    i += 1

0 	 id 	 11156
1 	 listing_url 	 https://www.airbnb.com/rooms/11156
2 	 scrape_id 	 20211106153124
3 	 last_scraped 	 2021-11-07 00:00:00
4 	 name 	 An Oasis in the City
5 	 description 	 Very central to the city which can be reached by an easy walk or by bus, with transport at the door, if required, and all amenities within easy reach.<br /><br /><b>The space</b><br />Potts Pt. is a vibrant  and popular inner-city suburb & the area was described as 'Australia's Bohemian Heart' in the article by Raymond Bonner in the New York Times . Ultra-convenient & within walking distance of the central business district  & all tourist sites such as: <br />the Botanic Gardens, <br />Opera House, <br />NSW Art Gallery, <br />Australian Museum,  <br />Circular Quay, the Sydney Opera House, and BridgeClimb<br />Darling Harbour<br />Museum of Contemporary Art<br />Art Gallery of NSW<br />Sydney Museum<br />Hyde Park<br />Botanic Gardens<br />Chinatown, Paddy's Markets, and the Chinese Gardens<br />Paddington (& the Saturday markets)<br />The historic Rocks area (& the markets on Saturday & Sunday)<br /><br />It is 7 minutes walk to the Kings Cross.train (Metro) station & buses to the be
6 	 neighborhood_overview 	 It is very close to everything and everywhere, has  many trees along the streets and is in a beautiful part of the city
7 	 picture_url 	 https://a0.muscache.com/pictures/2797669/17895d03_original.jpg
8 	 host_id 	 40855
9 	 host_url 	 https://www.airbnb.com/users/show/40855
10 	 host_name 	 Colleen
11 	 host_since 	 2009-09-23 00:00:00
12 	 host_location 	 Potts Point, New South Wales, Australia
13 	 host_about 	 Recently retired, I've lived & worked on 4 continents & over the years have travelled thru about 50 countries. I live in inner-Sydney within walking distance of most places (except the beaches, which are approx. 30 minutes. away on public transport).  Interests are travelling, wildlife, reading,snorkelling, theatre, ballet & interesting food. 
14 	 host_response_time 	 nan
15 	 host_response_rate 	 nan
16 	 host_acceptance_rate 	 nan
17 	 host_is_superhost 	 f
18 	 host_thumbnail_url 	 https://a0.muscache.com/im/users/40855/profile_pic/1259121939/original.jpg?aki_policy=profile_small
19 	 host_picture_url 	 https://a0.muscache.com/im/users/40855/profile_pic/1259121939/original.jpg?aki_policy=profile_x_medium
20 	 host_neighbourhood 	 Potts Point
21 	 host_listings_count 	 1.0
22 	 host_total_listings_count 	 1.0
23 	 host_verifications 	 ['email', 'phone', 'reviews']
24 	 host_has_profile_pic 	 t
25 	 host_identity_verified 	 f
26 	 neighbourhood 	 Potts Point, New South Wales, Australia
27 	 neighbourhood_cleansed 	 Sydney
28 	 neighbourhood_group_cleansed 	 nan
29 	 latitude 	 -33.86767
30 	 longitude 	 151.22497
31 	 property_type 	 Private room in rental unit
32 	 room_type 	 Private room
33 	 accommodates 	 1
34 	 bathrooms 	 nan
35 	 bathrooms_text 	 1 shared bath
36 	 bedrooms 	 1.0
37 	 beds 	 nan
38 	 amenities 	 ["TV", "Kitchen", "Refrigerator", "Extra pillows and blankets", "Cooking basics", "Stove", "Washer", "Dishwasher", "Hot water", "Dishes and silverware", "Essentials", "Backyard", "Oven", "Long term stays allowed", "Elevator", "Smoke alarm", "Single level home", "Hangers", "Shower gel", "Fire extinguisher", "Bed linens", "Wifi", "Microwave", "Hair dryer", "Shampoo", "Heating", "Iron", "Patio or balcony"]
39 	 price 	 $65.00
40 	 minimum_nights 	 90
41 	 maximum_nights 	 180
42 	 minimum_minimum_nights 	 90
43 	 maximum_minimum_nights 	 90
44 	 minimum_maximum_nights 	 180
45 	 maximum_maximum_nights 	 180
46 	 minimum_nights_avg_ntm 	 90.0
47 	 maximum_nights_avg_ntm 	 180.0
48 	 calendar_updated 	 nan
49 	 has_availability 	 t
50 	 availability_30 	 29
51 	 availability_60 	 59
52 	 availability_90 	 89
53 	 availability_365 	 364
54 	 calendar_last_scraped 	 2021-11-07 00:00:00
55 	 number_of_reviews 	 196
56 	 number_of_reviews_ltm 	 0
57 	 number_of_reviews_l30d 	 0
58 	 first_review 	 2013-11-12
59 	 last_review 	 2020-03-05
60 	 review_scores_rating 	 4.6
61 	 review_scores_accuracy 	 4.76
62 	 review_scores_cleanliness 	 4.31
63 	 review_scores_checkin 	 4.8
64 	 review_scores_communication 	 4.77
65 	 review_scores_location 	 4.82
66 	 review_scores_value 	 4.71
67 	 license 	 nan
68 	 instant_bookable 	 f
69 	 calculated_host_listings_count 	 1
70 	 calculated_host_listings_count_entire_homes 	 0
71 	 calculated_host_listings_count_private_rooms 	 1
72 	 calculated_host_listings_count_shared_rooms 	 0
73 	 reviews_per_month 	 2.02

When the data were collected?

df_listings_detailed['last_scraped'].value_counts()

2021-11-07    12543
2021-11-06     7970
Name: last_scraped, dtype: int64

How many properties and hosts are present in the dataset?

print('Number of properties:', df_listings_detailed['id'].unique().shape[0])
print('Number of hosts:', df_listings_detailed['host_id'].unique().shape[0])

Number of properties: 20513
Number of hosts: 14977

How many properties a host have in the list?

host_property_number = df_listings_detailed.groupby('host_id')['id'].nunique().value_counts()
host_property_number = host_property_number.reset_index()

host_property_number.columns = ["host_listing_number", "host_number"]
#host_property_number
print('The least number of properties listed by one host:', host_property_number.min()[0])
print('The largets number of proeprties list by one host:', host_property_number.max()[0])
print('In average, one host lists properties:', host_property_number.mean()[0])
print('Median number of properties listed by one host:', host_property_number.median()[0])

fig = px.bar(host_property_number,
             x='host_listing_number',
             y='host_number',
             labels={'host_listing_number': 'Number of Properties Listed by Same Host',
                     'host_number': 'Number of Hosts (log)' 
                    },
             log_y=True,
             title="NSW AirBnb Hosts' Listing Numbers "
            )
fig.show()

The least number of properties listed by one host: 1
The largets number of proeprties list by one host: 188
In average, one host lists properties: 33.07142857142857
Median number of properties listed by one host: 22.5

Diversities of accommodation types.

df_property_types = df_listings_detailed['property_type'].value_counts()
df_property_types = df_property_types.reset_index()
df_property_types.columns = ['property_type', 'number']

print("In the dateset, {} types of accommodations are listed.".format(df_property_types.shape[0]))
print("The largest number of accommodations are common types such as unit, room in unit, residential home.")
print("However, there are small numbers of unusual accommodation types e.g. earth house, tent.")
df_property_types
fig = px.pie(df_property_types,
             names='property_type',
             values='number',
             
             title="AirBnb Property Types"
            )
fig.update_traces(hoverinfo='label+percent+name', textinfo='none')
fig.update_layout(width=800,
                  height=800
                 )
fig.show()


df_room_types = df_listings_detailed['room_type'].value_counts()
df_room_types = df_room_types.reset_index()
df_room_types.columns = ['room_type', 'number']

print("Room types are rather limited. There are {} types of rooms listed.".format(df_room_types.shape[0]))
print("Majority of listings are made up of entire home/apt and private room.")

df_room_types
fig = px.pie(df_room_types,
             names='room_type',
             values='number',
             title="AirBnb Room Types"
            )
fig.update_traces(hoverinfo='label+percent+name')
fig.update_layout(width=800,
                  height=800
                 )
fig.show()


df_listings_detailed['price_in_number'] = df_listings_detailed['price'].str.replace('$', '')
df_listings_detailed['price_in_number'] = df_listings_detailed['price_in_number'].str.replace(',','')
df_listings_detailed['price_in_number'] = df_listings_detailed['price_in_number'].astype(float).astype(int)
fig = px.violin(df_listings_detailed, 
                y="price_in_number", 
                x="room_type", 
                color="room_type", 
                labels={'room_type': 'Room Type',
                        'price_in_number': 'Price'},
                box=False, 
                points='outliers',#points="all",
                hover_data=['price'],
                #width=2000,
                #height=2000,
                #log_y=True,
                title='Price of different types of rooms'
               )
fig.show()

In the dateset, 83 types of accommodations are listed.
The largest number of accommodations are common types such as unit, room in unit, residential home.
However, there are small numbers of unusual accommodation types e.g. earth house, tent.

Room types are rather limited. There are 4 types of rooms listed.
Majority of listings are made up of entire home/apt and private room.

<ipython-input-89-ca70f7a61bb2>:42: FutureWarning:

The default value of regex will change from True to False in a future version. In addition, single character regular expressions will *not* be treated as literal strings when regex=True.

Locations: number of properties and their prices in neighbourhoods.

df_listings_number_neighbourhood = df_listings_detailed.groupby('neighbourhood_cleansed')['id'].nunique()
df_listings_number_neighbourhood = df_listings_number_neighbourhood.reset_index()
df_listings_number_neighbourhood.columns = ['neighbourhood', 'property_number']
#df_listings_number_neighbourhood


fig = px.bar(df_listings_number_neighbourhood,
             x='neighbourhood', 
             y='property_number',
             hover_data=['neighbourhood', 'property_number'],
             labels={'neighbourhood': 'Neighbourhood', 'property_number': 'Number of Properties'},
             title='Number of properties in neighbourhoods',
             color='neighbourhood',
             log_y=True,
             height=400
            )
fig.show()


fig = px.violin(df_listings_detailed.sort_values(by=['neighbourhood_cleansed']), 
                y="price_in_number", 
                x="neighbourhood_cleansed", 
                color="neighbourhood_cleansed", 
                labels={'neighbourhood_cleansed': 'Neighbourhood',
                        'price_in_number': 'Price'},
                box=False, 
                points='outliers',#points="all",
                hover_data=['price'],
                #width=2000,
                height=400,
                log_y=True,
                title='Prices in Neighbourhoods'
               )
fig.show()

How long have the hosts been in the business?

print('Earlist host:', df_listings_detailed['host_since'].min())
print('Latest host:', df_listings_detailed['host_since'].max())
df_listings_detailed['host_since_date'] = df_listings_detailed['host_since'].dt.year
df_enter_business_date = df_listings_detailed['host_since_date'].value_counts().reset_index()
df_enter_business_date.columns = ['enter_business_date', 'number']
df_enter_business_date = df_enter_business_date.sort_values(by=['enter_business_date'])
df_enter_business_date
fig = px.bar(df_enter_business_date,
             x='enter_business_date',
             y='number',
             labels={'enter_business_date': 'Enter Business Date',
                     'number': 'Number of Hosts' 
                    },
             #log_y=True,
             title="Hosts Enter Buisiness Date"
            )
fig.show()

Earlist host 2009-03-20 00:00:00
Latest host 2021-11-04 00:00:00